Project Summary Vitamin D is a promising nutritional cancer prevention agent, but most of the relevant research has been conducted on breast, colorectal and prostate cancer. Less is known about the association of vitamin D with lung cancer risk. A major challenge of lung cancer prevention research is that smoking remains the strongest risk factor and has the potential for residual confounding in analyses that examine diet and other exposures. The Women's Health Initiative (WHI) provides an excellent environment in which to examine lung cancer risk factors other than smoking since smoking prevalence was low and a large number of the lung cancer cases are among never smokers. Therefore, our primary objective is to determine the association of vitamin D status with lung cancer risk in never-smoking, postmenopausal women in the WHI. In a nested case-control study of 300 incident lung cancer cases and 300 controls, we will test whether high vs. low serum concentrations of 25- hydroxyvitamin D [25(OH)D], a biomarker of vitamin D status, are associated with reduced lung cancer risk among never-smoking women. In addition, we hypothesize that vitamin A may modify the association of serum 25(OH)D with lung cancer risk. In nuclei, high concentrations of 9-cis-retinoic acid, an active metabolite of vitamin A and the ligand of retinoid X receptor, impedes the proper function of the vitamin D receptor. Thus, high intake of vitamin A, such as supplemental vitamin A use, likely leads to excessively high 9-cis-retinoic acid concentrations both in the circulation and in cells. Previous literature has not considered this potentially important nutrient-nutrient interaction. From our preliminary data, serum concentrations of 25(OH)D are inversely associated with lung cancer mortality among former and never smokers; the association may be attenuated by excess circulating vitamin A or vitamin A supplement use. Thus, in the secondary aim, we will test whether the reduced lung cancer risks associated with high serum concentrations of 25(OH)D are stronger among women without excess circulating vitamin A compared to those with excess circulating vitamin A. The excess circulating vitamin A is defined as serum retinyl esters e7 g/mL and the ratio of retinyl esters to retinol e0.08. We will use a chemiluminescent immunoassay to measure 25(OH)D and high-performance liquid chromatography with UV detector to measure retinyl esters and retinol in pre-diagnostic sera. Serum 25(OH)D concentrations will be classified by using season-specific quartiles, season-standardized quartiles, and clinically relevant concentrations. Logistic regression will be used to estimate multivariate-adjusted odds ratios and 95% confidence intervals for the associations of serum 25(OH) and lung cancer risk. The interaction of serum 25(OH)D and excess vitamin A on lung cancer risk will also be evaluated in regression models.